Overview
|
Over time, I've become much more interested in teaching; particularly at the masters level.
I have taught and developed a wide variety of classes, ranging from undergraduate to advanced PhD levels
at a variety of universities. See below for a list of classes along with
a good amount of course materials I have produced.
In my research, I have done work in several applied and theoretical areas.
I have worked extensively on developing both methods and theory
for solving various problems in
astronomy and cosmology. This research introduced me to the field known as
inverse problems, where the observed data are actually noisy version
of smooth functionals of the
object of interest.
Additionally, I have spent a fair amount of time researching the prediction risk
implications of empirical tuning parameter selection for lasso-type methods.
More recently, I have become interested in addressing some of the philosophies
that currently dominate the field of macroeconomic forecasting. Most notably,
the overparameterization and complexity that results the over reliance on microeconomic foundations
for doing predictions. Lastly, I have worked on examining the statistical implications of computational
approximations. More specifically, investigating the intriguing possibility
that these approximations can actually improve statistical performance.
For a cv, click
here.
|
Bio
|
I have a bachelors degree from the University of Colorado in
economics and math, and a masters and Ph.D. from Carnegie Mellon
University in statistics, under the direction
of Chris Genovese. Outside of academia, I enjoy welding and metal working, riding my bike, and,
perhaps most of all, coffee.
|